Nature © Macmillan Publishers Ltd 1998

8

letters to nature

NATURE | VOL 395 | 8 OCTOBER 1998 | www.nature.com 615

12. Bicknell, D. C. & Gower, D. B. The development and application of a radioimmunoassay for 5alpha-androst-16-en-3alpha-ol in plasma. J. Steroid Biochem. 7, 451–455 (1976).

13. Gower, D. B. et al. Comparison of 16-androstene steroid concentrations in sterile apocrine sweat andaxillary secretions: interconversions of 16-androstenes by the axillary microflora—a mechanism foraxillary odour production in man? J. Steroid Biochem. Mol. Biol. 48, 409–418 (1994).

14. Louveau, I., Bonneau, M. & Gower, D. B. Biosynthesis of 16-androstene steroids and testosterone byporcine testis tissue in vitro: effect of age and relationships with fat 5 alpha-androstenone levels in vivo.Acta Endocrinol. 125, 526–531 (1991).

15. Janowski, B. A. et al. An oxysterol signalling pathway mediated by the nuclear receptor LXR alpha.Nature 383, 728–731 (1996).

16. Lehmann, J. M. et al. Activation of nuclear receptor LXR by oxysterols defines a new hormoneresponse pathway. J. Biol. Chem. 272, 3137–3140 (1997).

17. Lala, D. S. et al. Activation of the orphan nuclear receptor steroidogenic factor 1 by oxysterols. Proc.Natl Acad. Sci. USA 94, 4895–4900 (1997).

18. Forman, B. M., Chen, J. & Evans, R. M. Hypolipidemic drugs, polyunsaturated fatty acids andeicosanoids are ligands for peroxisome proliferator-activated receptors a and d. Proc. Natl Acad. Sci.USA 94, 4312–4317 (1997).

19. Kliewer, S. A. et al. Fatty acids and eicosanoids regulate gene expression through direct interactionswith peroxisome proliferator-activated receptors a and g. Proc. Natl Acad. Sci. USA 94, 4318–4323(1997).

20. Forman, B. M., Umesono, K., Chen, J. & Evans, R. M. Unique response pathways are established byallosteric interactions among nuclear hormone receptors. Cell 81, 541–550 (1995).

21. Forman, B. M. et al. Identification of a nuclear receptor that is activated by farnesol metabolites. Cell81, 687–693 (1995).

22. Seol, W., Choi, H. S. & Moore, D. D. An orphan nuclear hormone receptor that lacks a DNA bindingdomain and heterodimerizes with other receptors. Science 272, 1136–1139 (1996).

23. Ladias, J. A. & Karathanasis, S. K. Regulation of the apolipoprotein AI gene by ARP-1, a novel memberof the steroid recetor superfamily. Science 251, 561–565 (1991).

24. Golemis, E. A. & Brent, R. Fused protein domains inhibit DNA binding by LexA. Mol. Cell. Biol. 12,3006–3014 (1992).

25. Gyuris, J., Golemis, E., Chartkov, H. & Brent, R. Cdi1, a human G1 and S phase protein phosphatasethat associates with Cdk2. Cell 75, 791–803 (1993).

26. Lee, J. W., Moore, D. D. & Heyman, R. A. A chimeric thyroid hormone receptor constitutively boundto DNA requires retinoid X receptor for hormone-dependent transcriptional activation in yeast. Mol.Endocrinol. 8, 1245–1252 (1994).

Acknowledgements. We thank D. Russell and M. Mahendroo for discussions during this work;R. Miesfeld for an expression vector for the human androgen receptor; and S. Kliewer and I. Schulmanfor SRC-1 plasmids. This work was supported by the Howard Hughes Medical Institute (R.M.E.), Marchof Dimes (R.M.E.), CapCURE (R.M.E.), Tobacco-Related Disease Research Program (B.M.F.), The City ofHope National Medical Center/Beckman Research Institute (B.M.F.), and National Institutes of Health(D.D.M.). R.M.E. is an Investigator of the Howard Hughes Medical Institute at the Salk Institute forBiological Studies.

Correspondence and requests for materials should be addressed to B.M.F. (e-mail: bforman@gte.net).

AMec1-andRad53-dependent checkpointcontrols late-firingoriginsofDNAreplicationCorrado Santocanale & John F. X. Diffley

Imperial Cancer Research Fund, Clare Hall Laboratories, South Mimms,Hertfordshire EN6 3LD, UK. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DNA replication in eukaryotic cells initiates from many replica-tion origins1 which fire throughout the S phase of the cell cycle in apredictable pattern: some origins fire early, others late2. Little isknown about how the initiation of DNA replication and theelongation of newly synthesized DNA strands are coordinatedduring S phase. Here we show that, in budding yeast,hydroxyurea, which blocks the progression of replication forksfrom early-firing origins, also inhibits the firing of late origins.These late origins are maintained in the initiation-competentprereplicative state for extended periods. The block to late originfiring is an active process and is defective in yeast with mutationsin the rad53 and mec1 checkpoint genes, indicating that regula-tion of late origin firing may also be an important component ofthe ‘intra-S-phase’ checkpoint3 and may aid cell survival underadverse conditions.

ARS305 is an efficient, early-firing replication origin4. Density-transfer experiments have shown that ARS305 fires efficiently whenS phase is interrupted with the ribonucleotide reductase inhibitorhydroxyurea5. Under these conditions, forks from ARS305 stallsomewhere within roughly 10 kilobases (kb) of the origin. Weused two methods to examine replication intermediates generated

by these stalled forks. Either digestion of purified genomic DNAwith single-strand-specific nuclease followed by neutral agarose gelelectrophoresis, or denaturing agarose gel electrophoresis of undi-gested genomic DNA, releases replication intermediates as smallfragments, whereas unreplicated genomic DNA (from the firsttechnique) or parental DNA (from the second), which are muchlarger, do not run into the gel (Fig. 1a). The position of individualsequences can be visualized by hybridization with specific DNAprobes. We have confirmed our major results using both assays.

In Fig. 1b, cells were first synchronized in G1 phase by using a-factor mating pheromone, and were then released into mediumcontaining hydroxyurea. Replication intermediates from ARS305are not present in a-factor-blocked cells (Fig. 1b, lane 1); they firstappear 30 min after release into hydroxyurea and persist for at least90 min, very slowly increasing in size to an average of roughly 4–5 kb. This is consistent with the amount of replication from ARS305seen in density-substitution experiments5. Thus, hydroxyurea doesnot inhibit initiation from early-firing origins but greatly slows theprogression of replication forks after initiation. The accumulationof replication intermediates in hydroxyurea is origin-specific, asreplication intermediates were not detected from a sequence, R11,that does not contain an origin (Fig. 1c). Most important, replica-tion intermediates were not detected from an efficient late-firingorigin, ARS501, even after long periods of time (Fig. 1c).

Cdc6p-dependent prereplicative complexes seem to be essentialin the initiation of DNA replication. Genomic footprinting experi-ments have indicated that initiation is accompanied by loss ofprereplicative complexes from origins and that new prereplicativecomplexes cannot assemble at origins until passage of cells throughmitosis6–9. The origin ARS1, in its normal location on chromosomeIV (ARS1IV), like ARS305, fires early in S phase4. The postreplicativestate of ARS1IV in cells blocked in G2/M phase with nocodazole6,8 ischaracterized primarily by the origin recognition complex (ORC)-induced DNase1-hypersensitive site (Fig. 2a, lanes 4–6, asterisk).Lanes 7–9 show the prereplicative state of ARS1IV in cells blocked inG1 with a-factor; this state is characterized by suppression of theORC-induced hypersensitive site and protection across the B1 andB2 elements. The final three lanes (10–12) show that when cells werereleased from the G1 block into hydroxyurea, ARS1IV reverted to thepostreplicative state. The same result was obtained for ARS305 (datanot shown). This is consistent with the finding that hydroxyureadoes not block the activation of these early-firing origins.

Placing ARS1 near a telomere forces it to fire later in S phase10.Therefore, we studied the effect of hydroxyurea on ARS1 in the late-firing context of a telomeric linear minichromosome (ARS1TEL).ARS1TEL forms a postreplicative complex in nocodazole-blockedcells (Fig. 2b; lanes 1, 2) and a prereplicative complex in a-factor-blocked cells (Fig. 2b, lanes 3, 4); these complexes are indistinguish-able from those formed at ARS1IV. In contrast to ARS1IV, however,ARS1TEL remains prereplicative after release into hydroxyurea for atleast 90 min (compare Fig. 2a and b). Therefore, conversion of ARS1to a late-firing origin renders its firing sensitive to inhibition byhydroxyurea. Consequently, the inhibition of origin firing byhydroxyurea does not appear to be an intrinsic property of specificorigin sequences but is dictated by their time of firing during Sphase.

ARS301 is inactive in its normal chromosomal location but active.95% of the time when present on a plasmid8,11,12. On a plasmid,ARS301 is activated after ARS305 (ref. 5). Density-substitutionexperiments have shown that there is no detectable DNA replicationfrom this plasmid in hydroxyurea5. When we used the DNA samplesfrom Fig. 2a to study the status of plasmid-borne ARS301, we foundthat, unlike ARS1IV and ARS305, but like ARS1TEL, ARS301remained prereplicative after release into hydroxyurea as shownby continued suppression of the ORC-induced hypersensitive sites(asterisks, Fig. 2c). Plasmid-borne ARS301 reverts to the post-replicative complex, and the plasmid duplicates in a Cdc7-depen-

Nature © Macmillan Publishers Ltd 1998

8

letters to nature

616 NATURE | VOL 395 | 8 OCTOBER 1998 | www.nature.com

8.4 kb7.26.45.74.84.33.72.31.9

1.31.4

0 15 30 45 60 75 90

α HU

Minutes

1 2 3 4 5 6 7

ARS305

305 501 R11

α HU

1 2 3 4 5 6

8.4 kb7.26.45.74.84.33.72.31.9

1.31.4

RI

HighMr

Mr

RI

a b c

High

Digestion of ssDNA with Mung Bean Nuclease

Neutral agarose gel electrophoresis

Alkaline agarose gel electrophoresis

Southern blot using ARS DNA as probe

Assay I Assay II

Replication bubble

ARS

ARS

α HU α HU

Figure 1 Analysis of replication intermediates (RIs) from stalled replication forks.

a, The assays used to detect RIs. Single-stranded (ss) DNA is indicated by the

short arrows. ARS, origin sequence. b, Detection of RIs from chromosomal

ARS305. a-factor-arrested cells (Y300 (ref. 23)) were released into medium

containing hydroxyurea (HU) for 0–90min. Replication intermediates from

ARS305 were studied using mung bean nuclease (see Methods). c, RI accumula-

tion is specific for an early-firing origin. At 0 and 90min after release of cells from

a-factor, samples were collected and processed as above. Probes for RI detec-

tion were specific for the early-firing origin ARS305 (305), the late-firing origin

ARS501 (501) and the non-origin-containing sequence R11.

Figure 2 Conversion from the pre- to the postreplicative state is blocked at late

firing origins of replication in hydroxyurea. a, Genomic footprinting analysis of

ARS1IV. Cells (W303-1a) carrying plasmid pCS1 (ref. 8) were arrested either in G2

phase with nocodazole (NOC) or in G1 phase with a-factor (a), or were released

from a-factor block into hydroxyurea for 90min (a → HU). ND, naked DNA.

Asterisk indicates ORC-induced hypersensitive sites and grey bar indicates G1-

specific protection. b, ARS1TEL firing is sensitive to inhibition by hydroxyurea.

Cells (W303-1a, DARS1IV::HIS3) carrying the linear telomeric plasmid pYLPV

(ref. 24) were treated as in a. Chromatin structure of plasmid-borne ARS1TEL

could be unambiguously detected because ARS1IV was deleted in this strain.

c, Activation of plasmid-borne ARS301 is blocked by hydroxyurea. DNA samples

from a were used to analyse the chromatin structure of ARS301 on plasmid pCS1.

a b c

Mr

RI

1C 2C

1C 2C

Figure 3 Inhibition of late origin firing is an active process. a, Bulk DNA replication

is blocked by hydroxyurea in both wild-type and rad53 mutant strains. DNA

content of wild-type (Y300 (ref. 23)) and rad53 mutant (Y301 (ref. 23)) cells released

from G1 into hydroxyurea-containing medium for 90min (black lines) was mea-

sured by flow cytometry. FACS profiles from logarithmically growing cells (grey

areas) are shown to indicate the positions of the 1C and 2C peaks. b, Plasmid-

borne ARS301 loses its prereplicative complex in rad53 mutant cells in

hydroxyurea rad53 mutant cells carrying plasmid pCS1 were treated as in

Fig. 2c. c, ARS501 fires in hydroxyurea in checkpoint-deficient strains. a-factor-

arrested cells (Y300, Y301, YMP10860, YMP10848 (ref. 25)) were released into

hydroxyurea-containing medium. RIs from ARS305 and ARS501 were studied

using the alkaline gel electrophoresis method.

Nature © Macmillan Publishers Ltd 1998

8

letters to nature

NATURE | VOL 395 | 8 OCTOBER 1998 | www.nature.com 617

dent reaction, after release from hydroxyurea5, showing thatARS301 remains competent to fire during the hydroxyurea block.These results show that firing of late origins is blocked in hydroxy-urea at the conversion from the pre- to the postreplicative complex.

Hydroxyurea activates a checkpoint that delays entry into mitosisuntil S phase is completed13. We therefore studied known check-point mutants for defects in the dependency of late origin firing onearly replication. rad9, mec3 or tel1 deletion mutants were all stillcapable of blocking ARS301 activation in hydroxyurea (data notshown); however, the mec1-1 mutant and two different rad53mutant alleles (also known as sad1, mec2 and spk1) were defectivein this dependency (Fig. 3; data not shown). After release from thea-factor block into hydroxyurea, rad53 mutant cells, like wild-typecells, do not replicate significant amounts of DNA, arresting with anear 1C DNA content (Fig. 3a). Thus, the block to bulk DNAsynthesis by hydroxyurea does not depend on Rad53. However,unlike wild-type cells, ARS301 is converted to the postreplicativestate in rad53 (compare Fig. 2c and Fig. 3b) and mec1 (data notshown) mutants after release into hydroxyurea. This is because lateorigins fire inappropriately in these checkpoint mutants (Fig. 3c).Replication intermediates from ARS305 can be detected in bothwild-type cells and rad53 mutants blocked in hydroxyurea (Fig. 3c,lanes 1–8), indicating that ARS305 is activated and that replicationforks stall in both strains. However, replication intermediates fromARS501 are not detected in wild-type cells blocked in hydroxyurea(Fig. 1c, lanes 9–12 and Fig. 3c, lanes 9–12, 17–20) whereas they canclearly be detected from ARS501 in either rad53 (lanes 13–16) ormec1 (lanes 21–24) mutant cells blocked in hydroxyurea. Replica-tion intermediates were not detected from non-origin-containingsequences such as R11 in these experiments (Supplementary infor-mation), indicating that accumulation of replication intermediatesis origin-specific even in the checkpoint mutants. These resultsshow that Rad53 and Mec1 are required to block the activation oflate-firing origins in hydroxyurea.

ARS501 replication intermediates appear 30 min later thanARS305 in both mutants (Fig. 3c), indicating that the relativeorder of origin firing is retained. We have also found that ARS501still fires after ARS305 in rad53 mutants in an unperturbed S phase,although its firing was slightly accelerated relative to ARS305(Supplementary information). It is likely that establishment of

replication forks from early-firing origins is required so thathydroxyurea can generate the signals that activate the checkpointpathway. Thus the Mec1/Rad53 pathway is unlikely to be solelyresponsible for establishing the normal order of origin firing.

We propose the model shown in Fig. 4. During G1 phase,prereplicative complexes assemble at future replication origins.This is roughly the time at which origins become committed toeither early or late firing in the subsequent S phase14. Early originsare preferentially activated by S-phase-promoting factors, such asthe protein kinases Cdc28 and Cdc7, by an unknown mechanism. Ifreplication forks stall, the firing of late origins is blocked by theMec1/Rad53 protein kinases. Given the homology between Mec1and the human ataxia telangiectasia mutated (ATM) protein15, wesuggest that the regulation of late origin firing described here may berelated to the fact that g-irradiation blocks activation of newreplicons in mammalian cells at an undefined step and that thisblock is defective in ATM mutant cell lines16,17. We also suggest thatthe Mec1/Rad53-dependent inhibition of late origin firing may helpto explain the expansion of S phase when DNA has been damaged bymethylating agents3. Further characterization of this ‘origin-firingcheckpoint’ should help in determining how S-phase events arecoordinated in eukaryotic cells and how anti-cancer drugs thattarget DNA replication, such as hydroxyurea, work. M. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Methods

Cell-cycle arrests, fluorescence-activated cell sorting (FACS) and genomicfootprinting were performed as described6,8,18–20. Hydroxyurea was used at0.2 M at 25 8C in all experiments. For the analysis of replication intermediates,genomic DNA was prepared as described5. DNA prepared from 107 cells wasdigested for 1 h at 30 8C with 10 U of mung bean nuclease (New EnglandBiolabs) and separated on 0.8% neutral gels in Tris–acetate-EDTA buffer21.Alkaline gel electrophoresis was performed using 1% agarose gels asdescribed21. DNA was transferred from gels to Hybond N+ membranesaccording to the manufacturer’s instructions. Probes for ARS305 origin andR11 sequence on chromosome V have been described4,22. A probe specific forthe late-replicating origin ARS501 was amplified by the polymerase chainreaction from genomic DNA using the oligonucleotides 59-GGA TCC CGAGTC ATG TTT GG-39 and 59-GAG CAT AAT TAT GAC TGT AGC CC-39.

Received 16 March; accepted 15 July 1995.

1. Huberman, J. A. & Riggs, A. D. On the mechanism of DNA replication in mammalian chromosomes.J. Mol. Biol. 32, 327–341 (1968).

2. Fangman, W. L. & Brewer, B. J. A question of time–replication origins of eukaryotic chromosomes.Cell 71, 363–366 (1992).

3. Paulovich, A. G. & Hartwell, L. H. A checkpoint regulates the rate of progression through S phase in S.cerevisiae in response to DNA damage. Cell 82, 841–847 (1995).

4. Reynolds, A. E., McCarroll, R. M., Newlon, C. S. & Fangman, W. L. Time of replication of ARSelements along yeast chromosome III. Mol. Cell. Biol. 9, 4488–4494 (1989).

5. Bousset, K. & Diffley, J. F. X. The Cdc7 protein kinase is required for origin firing during S phase. GenesDev. 12, 480–490 (1998).

6. Diffley, J. F. X., Cocker, J. H., Dowell, S. J. & Rowley, A. Two steps in the assembly of complexes at yeastreplication origins in vivo. Cell 78, 303–316 (1994).

7. Cocker, J. H., Piatti, S., Santocanale, C., Nasmyth, K. & Diffley, J. F. X. An essential role for the Cdc6protein in forming the pre-replicative complexes of budding yeast. Nature 379, 180–182 (1996).

8. Santocanale, C. & Diffley, J. F. X. ORC- and Cdc6-dependent complexes at active and inactivechromosomal replication origins in Saccharomyces cerevisiae. EMBO J. 15, 6671–679 (1996).

9. Piatti, S., Bohm, T., Cocker, J. H., Diffley, J. F. X. & Nasmyth, K. Activation of S-phase promotingCDKs in late G1 defines a ‘‘point of no return’’ after which Cdc6 synthesis cannot promote DNAreplication in yeast. Genes Dev. 10, 1516–1531 (1996).

10. Ferguson, B. M. & Fangman, W. L. A position effect on the time of replication origin activation inyeast. Cell 68, 333–339 (1992).

11. Dubey, D. D. et al. Evidence suggesting that the ARS elements associated with silencers of the yeastmating-type locus HML do not function as chromosomal DNA replication origins. Mol. Cell. Biol. 11,5346–5355 (1991).

12. Broach, J. R. et al. Localization and sequence analysis of yeast origins of DNA replication. Cold SpringHarb. Symp. Quant. Biol. 47, 1165–1173 (1982).

13. Elledge, S. J. Cell cycle checkpoints: preventing an identity crisis. Science 274, 1664–1672 (1996).14. Raghuraman, M. K., Brewer, B. J. & Fangman, W. L. Cell cycle-dependent establishment of a late

replication program. Science 276, 806–809 (1997).15. Zakian, V. A. ATM-related genes: what do they tell us about functions of the human gene? Cell 82,

685–687 (1995).16. Painter, R. B. & Young, B. R. Radiosensitivity in ataxia-telangiectasia: a new explanation. Proc. Natl

Acad. Sci. USA 77, 7315–7317 (1980).17. Larner, J. M., Lee, H. & Hamlin, J. L. Radiation effects on DNA synthesis in a defined chromosomal

replicon. Mol. Cell. Biol. 14, 1901–1908 (1994).18. Santocanale, C., Neecke, H., Longhese, M. P., Lucchini, G. & Plevani, P. Mutations in the gene

encoding the 34 kDa subunit of yeast replication protein A cause defective S phase progression. J. Mol.Biol. 254, 595–607 (1995).

SPF

Rad53Mec1

L

L

E L

Figure 4 Model for regulation of late-firing origins. During G1 phase,

prereplicative complexes (diamonds) assemble at both early (E)- and late (L)-

firing origins of replication. Early origins are preferentially activated by S-phase-

promoting factors (SPF). If replication forks (triangles) stall (black rectangles), the

firing of late origins is blocked by the activity of Mec1 andRad53. Arrows from SPF

indicate the relative activity of SPF towards early and late origins.

Nature © Macmillan Publishers Ltd 1998

8

letters to nature

618 NATURE | VOL 395 | 8 OCTOBER 1998 | www.nature.com

19. Rowley, A., Cocker, J. H., Harwood, J. & Diffley, J. F. X. Initiation complex assembly at budding yeastreplication origins begins with the recognition of a bipartite sequence by limiting amounts of theinitiator, ORC. EMBO J. 14, 2631–2641 (1995).

20. Diffley, J. F. X. & Cocker, J. H. Protein–DNA interactions at a yeast replication origin. Nature 357,169–172 (1992).

21. Sambrook, J., Fritsch, E. F. & Maniatis, T. Molecular Cloning: a Laboratory Manual (Cold Spring Harb.Lab. Press, Cold Spring Harbor, 1989).

22. Ferguson, B. M., Brewer, B. J., Reynolds, A. E. & Fangman, W. L. A yeast origin of replication isactivated late in S phase. Cell 65, 507–515 (1991).

23. Allen, J. B., Zhou, Z., Siede, W., Friedberg, E. C. & Elledge, S. J. The SAD1/RAD53 protein kinasecontrols multiple checkpoints and DNA damage-induced transcription in yeast. Genes Dev. 8, 2401–2415 (1994).

24. Wellinger, R. J., Ethier, K., Labrecque, P. & Zakian, V. A. Evidence for a new step in telomeremaintenance. Cell 85, 423–433 (1996).

25. Paulovich, A. G., Margulies, R. U., Garvik, B. M. & Hartwell, L. H. RAD9, RAD17, and RAD24 arerequired for S phase regulation in Saccharomyces cerevisiae in response to DNA damage. Genetics 145,45–62 (1997).

Supplementary information is available on Nature’s World-Wide Web site (http://www.nature.com) oras paper copy from the London editorial office of Nature.

Acknowledgements. We thank S. Elledge and L. Hartwell for yeast strains; V. A. Zakian for plasmidpYLPV; members of the Diffley lab for stimulating discussions; and members of the ICRF Photography,Peptide Synthesis and Oligonucleotide Synthesis departments for their help.

Correspondence and requests for materials should be addressed to J.F.X.D. (J. Diffley@icrf.icnet.uk).

RegulationofDNA-replicationoriginsduringcell-cycle progressionKatsuhiko Shirahige*, Yuji Hori*, Katsuya Shiraishi*,Minoru Yamashita†, Keiko Takahashi*, Chikashi Obuse*,Toshiki Tsurimoto* & Hiroshi Yoshikawa*

* Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma,Nara 630-0101, Japan† Ajinomoto Co. Inc., 1-1 Suzuki, Kawasaki, Kawasaki City 210-8680, Japan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

We have shown previously that chromosome VI of Saccharomycescerevisiae contains nine origins of DNA replication that differ ininitiation frequency and replicate sequentially during the S phaseof the cell cycle1,2. Here we show that there are links betweenactivation of these multiple origins and regulation of S-phaseprogression. We study the effects of a DNA-damaging agent,methyl methane sulphonate (MMS), and of mutations in check-point genes such as rad53 (ref. 3) on the activity of origins,measured by two-dimensional gel analysis, and on cell-cycleprogression, measured by fluorescence-activated cell sorting. Wefind that when MMS slows down S-phase progression it alsoselectively blocks initiation from late origins. A rad53 mutationenhances late and/or inefficient origins and releases the initiationblock by MMS. Mutation of rad53 also results in a late originbecoming early replicating. We conclude that rad53 regulates thetiming of initiation of replication from late origins during normalcell growth and blocks initiation from late origins in MMS-treatedcells. rad53 is, therefore, involved in the cell’s surveillance of S-phase progression4,5. We also find that orc2, which encodes sub-unit 2 of the origin-recognition complex6,7, is involved in suppres-sion of late origins.

Among four origins on the right arm of chromosome VI, ori607initiated replication immediately early in S phase and a telomere-proximal origin, ori609, initiated 30 min later. Two other origins,ori606 and ori608 (an inefficient origin used in only 5% of the cellcycle), initiated 5–10 min later than ori607. Replication of the 100-kilobase (kb) telomere-proximal region of the right arm of chro-mosome VI was unusually slow and varied among different cells of apopulation. We thought that such a slow and heterogeneous rate ofreplication of this region of the chromosome might be due to thedownregulation of replication of this region in response to physio-logical variations of individual cells. To test this hypothesis, we

studied the effect of MMS on the replication of the four originregions on the right arm of chromosome VI, because damage toDNA in general blocks S-phase progression3,8. We also studied theeffects of rad53 and mec1 mutations because the products of thesegenes are involved in the cell’s surveillance of S-phaseprogression3,4,8.

First, we studied the effect of MMS on cell-cycle progression andcell viability. In the wild-type cell, MMS blocked cell-cycle progres-sion during S phase without affecting viability, indicating that thecell’s surveillance mechanism is functioning in S phase in MMS-treated cells (Fig. 1a, d, f). In contrast, cells continued to proceedthrough S phase to G2 phase concomitantly with the loss of viabilityin rad53-1 and mec1-1 mutant cells (Fig. 1b, c, f )3. We also studiedorc2-1 mutant cells; the viability of these cells was sensitive to MMSat the permissive temperature (Fig. 1f ) and the cells were notblocked by MMS during S-phase progression (Fig. 1d, e). Toconfirm these results we studied the effect of MMS on the pro-gression through S phase of cells that had been arrested at G1 phase

Figure 1 Effect of the rad53-1, mec1-1 and orc2-1 mutations on cell-cycle

progression and viability of MMS-treated cells. Cells were grown in YPDA

medium to A600 ¼ 0:6 (0 h) and MMS (0.015%) was added. a–e, Cell-cycle dis-

tribution and f, cell viability were measured by FACS (a–e) and plating on YPDA (f)

at the indicated times after MMS addition. a–e,1C and 2C indicate cells with one

and two sets of chromosomes, respectively. a, Wild-type strain 7830-2-4A. b, c, Its

two mutants, rad53-1 (strain DLY264) (b) and mec1-1 (strain DLY285) (c), all grown

at 30 8C. d, Wild-type strain W303-1B. e, Its orc2-1 mutant strain (JRY4125) both

grown at 26 8C. f, Open squares, strain 7830-2-4A; filled squares, strain W303-1B;

filled circles, rad53-1 mutant; filled triangle, mec1-1 mutant; open circles, orc2-1

mutant.